Revealing native folding pathways for membrane proteins: My research group aims to understand how membrane proteins form their tertiary or quaternary structures. To answer this complex question, we employ a diverse array of single-molecule biophysics and cell biology tools to observe membrane protein folding in vitro as well as in cellular conditions, in unprecedented details. Before delving into this question, we have gained our experiences on how to apply single-molecule techniques to various membrane-involving biological processes. We have reported the disassembly mechanism of the SNARE complex by a proteasome system (20S complex) that provides important insights into how AAA+ ATPases tightly couple their ATP hydrolysis with unfolding of protein substrates (JACS 2013, Science 2015 and Nat. Comm. 2021). Our efforts finally culminate in a single-molecule approach based on magnetic tweezers, which permits observation of detailed folding pathways of multi-pass membrane proteins with a resolution of a few amino acids (Nat. Comm. 2013, 2014, 2018, Nat. Chem. Biol. 2015, Science 2019, and Nat. Chem. Biol. 2022). Our lab hopes to reveal the folding pathways of important human proteins, which may in turn shed lights into their evolutionary history, ER membrane protein homeostasis, and facilitation of folding of these membrane proteins for industrial applications.

 Development of Tools for Precision Medicine: My research group is also interested in development of new single-molecule imaging tools. We recently developed single-molecule version of co-immunoprecipitation (co-IP) analysis (Nat. Comm. 2013, Nat. Prot. 2013). The co-IP analysis has been used as the gold standard of determining protein-protein interactions (PPIs) in the modern molecular biology for several decades. By adopting single-molecule fluorescence microscopy as the detection method (instead of SDS PAGE-gel and western blotting), we have improved the sensitivity and time-resolution of the co-IP analysis by five orders of magnitude, respectively. With the extreme sensitivity and quantitativeness of the developed tool, we demonstrate development of PPI biomarkers that allow precision prediction of drug responses of individual tumors, in particular, in cancer types without actionable genomic mutations. This suggests a path toward “personalized diagnosis of cancers at the PPI level”, which would expand the concept and scope of the targeted cancer therapy (Nat. Biomed. Eng. 2018, eLife 2020).